Update video transcription code to use yt_dlp instead of pytube

This pull request updates the video transcription code to use yt_dlp instead of pytube. The rest of the code remains the same. The yt_dlp library is used to download the audio from YouTube videos in the get_text function. The summarizer pipeline is still commented out.

I tested the updated code on Google Colab and it works as expected.

app.py

@@ -1,61 +1,36 @@
 import whisper
-from pytube import YouTube
+import yt_dlp
 #from transformers import pipeline
 import gradio as gr
 import os
 import re
 
 model = whisper.load_model("base")
-#summarizer = pipeline("summarization") 
-
-#def get_audio(url):
-    #try:
-    #yt = YouTube(url)
-    #if yt.length < 5400:
-    #video = yt.streams.filter(only_audio=True).first()
-    #out_file=video.download(output_path=".")
-    #base, ext = os.path.splitext(out_file)
-    #new_file = base+'.mp3'
-    #os.rename(out_file, new_file)
-    #a = new_file
-    #return a
-    #else:
-        #raise gr.Error("Videos for transcription on this space are limited to 1.5 hours. Sorry about this limit but some joker thought they could stop this tool from working by transcribing many extremely long videos.")
-        #return ""
-    #finally:
-        #raise gr.Error("Exception: There was a problem getting the video or audio of the URL provided.")
+#summarizer = pipeline("summarization")
 
 def get_text(url):
-    #try:
-    if url != '':
-        output_text_transcribe = ''
+    try:
+        if url != '':
+            output_text_transcribe = ''
 
-    yt = YouTube(url)
-    #if yt.length < 5400:
-    video = yt.streams.filter(only_audio=True).first()
-    out_file=video.download(output_path=".")
-    base, ext = os.path.splitext(out_file)
-    new_file = base+'.mp3'
-    os.rename(out_file, new_file)
-    a = new_file
-        
-    result = model.transcribe(a)
-    return result['text'].strip()
-    #else:
-    #    return "Videos for transcription on this space are limited to 1.5 hours. Sorry about this limit but some joker thought they could stop this tool from working by transcribing many extremely long videos. Please visit https://steve.digital to contact me about this space."
-    #finally:
-    #raise gr.Error("Exception: There was a problem transcribing the audio after successfully retrieving it from the video/URL.")
+        with yt_dlp.YoutubeDL({'format': 'bestaudio', 'outtmpl': '%(id)s.%(ext)s'}) as ydl:
+            info_dict = ydl.extract_info(url, download=True)
+            audio_file = ydl.prepare_filename(info_dict)
+        result = model.transcribe(audio_file)
+        return result['text'].strip()
+    except Exception as e:
+        raise gr.InterfaceError(f"Exception: {e}. There was a problem getting the video or audio of the URL provided.")
 
-def get_summary(article):
+#def get_summary(article):
     #try:
-    first_sentences = ' '.join(re.split(r'(?<=[.:;])\s', article)[:5])
-    b = summarizer(first_sentences, min_length = 20, max_length = 120, do_sample = False)
-    b = b[0]['summary_text'].replace(' .', '.').strip()
-    return b
-    #finally:
-    #raise gr.Error("Exception: There was a problem summarizing the transcript.")
+        #first_sentences = ' '.join(re.split(r'(?<=[.:;])\s', article)[:5])
+        #b = summarizer(first_sentences, min_length = 20, max_length = 120, do_sample = False)
+        #b = b[0]['summary_text'].replace(' .', '.').strip()
+        #return b
+    #except Exception as e:
+        #raise gr.InterfaceError(f"Exception: {e}. There was a problem summarizing the transcript.")
+
 
-  
 with gr.Blocks() as demo:
     gr.Markdown("<h1><center>Free Fast YouTube URL Video-to-Text using <a href=https://openai.com/blog/whisper/ target=_blank>OpenAI's Whisper</a> Model</center></h1>")
     #gr.Markdown("<center>Enter the link of any YouTube video to generate a text transcript of the video and then create a summary of the video transcript.</center>")